{"id":"W2950704571","doi":"10.1109/icde.2019.00149","title":"Distributed Discovery of Functional Dependencies","year":2019,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Focus (optics); Distributed computing; Distributed database; Distributed algorithm; Computation; Distributed Computing Environment; Big data; Distributed data store; Data mining; Algorithm","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002882762,0.001012838,0.00175421,0.002303343,0.00123202,0.001735667,0.002809922,0.001080027,0.002706475],"category_scores_gemma":[0.01488184,0.0007524144,0.001282021,0.002800752,0.001279328,0.003727983,0.002653366,0.001671372,0.0005355855],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009764773,"about_ca_system_score_gemma":0.002972006,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003960059,"about_ca_topic_score_gemma":0.006532419,"domain_scores_codex":[0.9972446,0.0005706524,0.0001209712,0.0009411582,0.0008801277,0.0002424611],"domain_scores_gemma":[0.9877433,0.00664496,0.0008784432,0.003170759,0.001180156,0.0003824158],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005081821,0.0005215458,0.01835914,0.000731591,0.0003654698,0.0009770035,0.000482498,0.3992507,0.01203654,0.05937346,0.01651901,0.4908749],"study_design_scores_gemma":[0.00005104532,0.00005267452,0.001540104,0.00001567343,0.00004158683,0.0002263358,0.00007823342,0.8926124,0.003395138,0.09829146,0.003676377,0.00001903347],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03368293,0.0004157829,0.9611402,0.0006326312,0.00008258723,0.0001020059,0.0006444391,0.001353658,0.0019457],"genre_scores_gemma":[0.5560923,0.0003836237,0.4371466,0.0003080156,0.0002037796,0.0002256826,0.002516158,0.0002262356,0.002897605],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003960059,"threshold_uncertainty_score":0.01524568,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.148624552532601,"score_gpt":0.3635099975135841,"score_spread":0.2148854449809832,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}